Aruba Campus Wireless Networks

Aruba Campus Wireless Networks
Version 8


Validated Reference Design

Copyright
© 2011 Aruba Networks, Inc. AirWave®, Aruba Networks®, Aruba Mobility Management System®, Bluescanner, For Wireless That
Works®, Mobile Edge Architecture®, People Move. Networks Must Follow®, RFprotect®, The All Wireless Workplace Is Now Open For
Business, Green Island, and The Mobile Edge Company® are trademarks of Aruba Networks, Inc. All rights reserved. Aruba Networks
reserves the right to change, modify, transfer, or otherwise revise this publication and the product specifications without notice. While
Aruba uses commercially reasonable efforts to ensure the accuracy of the specifications contained in this document, Aruba will assume
no responsibility for any errors or omissions.

Open Source Code
Certain Aruba products include Open Source software code developed by third parties, including software code subject to the GNU
General Public License (“GPL”), GNU Lesser General Public License (“LGPL”), or other Open Source Licenses. The Open Source code
used can be found at this site:
http://www.arubanetworks.com/open_source

Legal Notice
ARUBA DISCLAIMS ANY AND ALL OTHER REPRESENTATIONS AND WARRANTIES, WEATHER EXPRESS, IMPLIED, OR
STATUTORY, INCLUDING WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE,
NONINFRINGEMENT, ACCURACY AND QUET ENJOYMENT. IN NO EVENT SHALL THE AGGREGATE LIABILITY OF ARUBA
EXCEED THE AMOUNTS ACUTALLY PAID TO ARUBA UNDER ANY APPLICABLE WRITTEN AGREEMENT OR FOR ARUBA
PRODUCTS OR SERVICES PURSHASED DIRECTLY FROM ARUBA, WHICHEVER IS LESS.

www.arubanetworks.com
1344 Crossman Avenue
Sunnyvale, California 94089
Phone: 408.227.4500
Fax 408.227.4550

Aruba Networks, Inc.

2



Table of Contents
Chapter 1:

Aruba Reference Architectures

7

Reference Material

8

Icons Used in this Guide

9

Campus Deployments

11

Aruba Campus WLAN Logical Architecture

12

Recommendations for Key Components

15

Master/Local Operation

19

Controller Licenses
Licensing Master Mobility Controllers
Licensing Local Mobility Controllers

19
20
20

Certificates

20

VLAN Design and Recommendations

23

VLAN Pooling

25

Redundancy

29

Master Redundancy

29

Local Redundancy

33

User Roles, Profiles, and AP Groups

39

Alias

40

Configuration Profiles

41

AP Groups

42

Chapter 7:

AP Groups for Client Access

43

Chapter 8:

Configuring the Employee Role

45

Configuring the Common Policy

45

Configuring the sip-session-allow Policy

47

Configuring the ocs-lync Policy

48


50

Employee VAP Profiles

53

Configuring the SSID Profiles
Configuring the Employee SSID Profile
Configuring Wi-Fi Multimedia

54
55
55

Chapter 2:

Chapter 3:

Chapter 4:

Chapter 5:

Chapter 6:

Chapter 9:


Table of Contents | 3



Configuring the AAA Profiles
Authentication Server and Server Groups
Configuring the NPS Server Group for 802.1X Authentication
Configuring the Employee AAA Profile
Configuring the Employee VAP Profiles

65
66
67

Configuring the Application SSID Profile

68

Configuring the Application AAA Profile

70

Configuring the Application VAP Profiles

71

Configuring the Guest Roles and VAP Profile

75

Configuring the Amigopod Policy

76

Configuring the guest-logon-access Policy

77

Configuring the block-internal-access Policy for the Guest Role

80

Configuring the auth-guest-access Policy

81

Configuring the drop-and-log Policy

82

Configuring the Initial Guest Role

83

Configuring the Authenticated Guest Role

85

Maximum User Sessions for Guest Role

86

Configuring the Guest SSID Profile

87

Configuring the Server Group for Guest Authentication

88

Configuring the Captive Portal Authentication Profile for Guest WLAN

89

Configuring the Guest AAA Profile

92

Configuring the Guest VAP Profile

93

Configuring the Radio Profiles

95

Configuring the ARM Profile

95

Configuring the 802.11a and 802.11g Radio Profiles

Chapter 12:

Configuring the Application Role and VAP Profiles
Configuring the Application Role

Chapter 11:

61

Configuring the tftp-session-allow Policy

Chapter 10:

57
57
58
59

98

Chapter 13:

Configuring the AP System Profiles

101

Chapter 14:

Configuring the QoS

105

Traffic Management Profile

105

VoIP Call Admission Control Profile

107





Chapter 15:

Configuring the Client Access AP Groups

109

Chapter 16:

AP Groups for Air Monitors

113

Configuring the AM Scanning Profile

114

Configuring the 802.11a and 802.11g Radio Profiles

116

Configuring the AP Groups for Air Monitors

118

Chapter 17:

Altering the Default AP Group for Pre 6.1 ArubaOS

119

Chapter 18:

Wireless Intrusion Prevention (IDS Profiles) of RFProtect

121

Chapter 19:

Spectrum Analysis

127

Chapter 20:

Mobility

131

Configuring the Mobility Domain

131

Chapter 21:

Control Plane Security

137

Chapter 22:

AP Provisioning

141

Chapter 23:

Logging

143

Chapter 24:

AirWave

145

Chapter 25:

Amigopod

147

Appendix A: Link Aggregation
Configuring LACP

Appendix B: Aruba Contact Information
Contacting Aruba Networks


149
149

155
155




Chapter 1: Aruba Reference Architectures
The Aruba Validated Reference Design (VRD) series is a collection of technology deployment guides
that include descriptions of Aruba technology, recommendations for product selections, network
design decisions, configuration procedures, and best practices for deployment. Together these guides
comprise a reference model for understanding Aruba technology and designs for common customer
deployment scenarios. Each Aruba VRD network design has been constructed in a lab environment
and thoroughly tested by Aruba engineers. Our customers use these proven designs to rapidly deploy
Aruba solutions in production with the assurance that they will perform and scale as expected.
The VRD series focuses on particular aspects of Aruba technologies and deployment models.
Together the guides provide a structured framework to understand and deploy Aruba wireless LANs
(WLANs). The VRD series has four types of guides:
 Foundation: These guides explain the core technologies of an Aruba WLAN. The guides also
describe different aspects of planning, operation, and troubleshooting deployments.
 Base Design: These guides describe the most common deployment models, recommendations,
and configurations.
 Applications: These guides are built on the base designs. These guides deliver specific
information that is relevant to deploying particular applications such as voice, video, or outdoor
campus extension.
 Specialty Deployments: These guides involve deployments in conditions that differ significantly
from the common base design deployment models, such as high-density WLAN deployments.

Specialty
Deployments

Applications

Foundation
Figure 1

arun_0334

Base Designs

VRD Core Technologies

This guide covers the deployment of Aruba WLAN in a typical campus network, and it is considered
part of the base designs guides within the VRD core technologies series. This guide covers the design
recommendations for a campus deployment and it explains the various configurations needed to
implement the Aruba secure, high-performance, multimedia grade WLAN solution in large campuses.
This guide describes these specific topics:
 recommended campus network design


Aruba Reference Architectures | 7









configuration of redundancy in campus deployments
configuration of AP groups for client access and air monitors
configuration of spectrum monitors (SMs)
configuration of Layer 3 mobility
configuration of control plane security (CPsec)

Table 1 lists the current software versions for this guide.
Table 1

Software Versions

Product

Version

ArubaOS (mobility controllers)

6.1

ArubaOS (mobility access switch)

7.0

ArubaInstant

1.1

MeshOS

4.2

AirWave

7.3

AmigopodOS

3.3

Reference Material
This guide is a base designs guide, and therefore it will not cover the fundamental wireless concepts.
This guide helps a wireless engineer configure and deploy the Aruba WLAN in a campus environment.
Readers should have a good understanding of wireless concepts and the Aruba technology that are
explained in the foundation-level guides.
 For information on indoor MIMO WLANs, see the Aruba 802.11n Networks Validated Reference
Design, available on the Aruba website at http://www.arubanetworks.com/vrd
 For information on Aruba Mobility Controllers and deployment models, see the Aruba Mobility
Controllers and Deployment Models Validated Reference Design, available on the Aruba
website at http://www.arubanetworks.com/vrd
 For specific deployment configuration details, or for deployment models for 802.11a/b/g
networks, see the 3.X series of VRDs on the Aruba website at http://www.arubanetworks.com/
vrd. The existing VRDs will be updated to follow this new format.
 The complete suite of Aruba technical documentation is available for download from the Aruba
support site. These documents present complete, detailed feature and functionality explanations
beyond the scope of the VRD series. The Aruba support site is located at: https://
support.arubanetworks.com/. This site requires a user login and is for current Aruba customers
with support contracts.




For more training on Aruba products or to learn about Aruba certifications, visit the Aruba
training and certification page on our website. This page contains links to class descriptions,
calendars, and test descriptions: http://www.arubanetworks.com/training.php/
Aruba hosts a user forum site and user meetings called Airheads. The forum contains
discussions of deployments, products, and troubleshooting tips. Airheads Online is an







invaluable resource that allows network administrators to interact with each other and Aruba
experts. Announcements for Airheads in person meetings are also available on the site: http://
airheads.arubanetworks.com/
The VRD series assumes a working knowledge of Wi-Fi®, and more specifically dependent AP,
or controller based, architectures. For more information about wireless technology
fundamentals, visit the Certified Wireless Network Professional (CWNP) site at http://
www.cwnp.com/

Icons Used in this Guide
Figure 2 shows the icons that are used in this guide to represent various components of the system.

Air
monitor

Spectrum
monitor

Router

Laptop

Microwave

Server

Mobile phone

Figure 2


Switch

AirWave
server

Firewall

Mobility
controller

Amigopod
server

Network cloud

arun_0332

AP

VRD Icon Set




Chapter 2: Campus Deployments
Campus deployments are networks that require more than a single active controller to cover a
contiguous space. Examples of campus deployments are corporate campuses, large hospitals, and
higher-education campuses. In these deployments, the WLAN is typically the primary access method
for the network, and it is typically used by multiple classes of users and devices. Figure 3 depicts a
cluster-based architecture that is typical of large enterprise deployments.

Data center
AirWave

Amigopod

Master
active

Master
standby

File
POS

PBX
RADIUS

Internet

Local
mobility controller

Air monitor

Figure 3


arun_0279

Local
mobility
controller

Typical campus deployment with redundancy

Campus Deployments | 11



Aruba WLAN has a logical four-tier operating model that consists of these four layers:
 Management: The management layer consists of AirWave®. AirWave provides a single point of
management for the WLAN, including reporting, heat maps, centralized configuration, and
troubleshooting.
 Network services: The network services layer consists of master mobility controllers and
Amigopod. Amigopod provides secure and flexible visitor management services. The master
controllers provide a control plane for the Aruba WLAN that spans the physical geography of the
wired network. The control plane does not directly deal with user traffic or APs. Instead the
control plane provides services such as whitelist coordination, valid AP lists, CPsec certificates,
RFProtect™ coordination, and RADIUS or AAA proxy.
 Aggregation: The aggregation layer is the interconnect point where the AP, air monitor (AM),
and SM traffic aggregates. This layer provides a logical point for enforcement of roles and
policies on centralized traffic that enters or exits the enterprise LAN.
 Network access: The network access layer is comprised of APs, AMs, and SMs that work
together with the aggregation layer controllers to overlay the Aruba WLAN.





Data center

Management

AirWave

File
Master
active

Network
services

Master
standby

POS

PBX
RADIUS

Amigopod

Control

Aggregation
Local
active

Local
active

Data
Retail_107

Network
access
Air
monitor

arun_0202

Air
monitor

Figure 4






An example network is used to explain the deployment of an Aruba user-centric network in the large
complex campus network presented in Chapter 2: Campus Deployments on page 11. All networks
parameters, screenshots, and command line interface (CLI) examples shown in this VRD are from the
VRD example network. For details on the network parameters, design and setup of the entire VRD
example network, see the Base Designs Lab Setup for Validated Reference Design.
Data center
AirWave

MC1-Sunnyvale3600 (active)

MC2-Sunnyvale3600 (standby)

SIP

Amigopod

Web
Radius
AD-CS
AD-DS
DNS
DHCP

Internet
Core switch
LC1Sunnyvale6000

LC2Sunnyvale6000

Distribution
switch 1

Distribution
switch 2

Access switch

AM-LC1

Employee

Guest
Application

Figure 5

AP-LC2

AM-LC2

Employee

Guest
Application

arun_0337

AP-LC1

Access switch

VRD example network

This VRD describes how to configure these WLANs:
 employee WLAN
 application WLAN (for 802.1X-incapable VoIP devices)
 guest WLAN
Employee WLAN emulates a converged voice and data network typical of most campus deployments.
Employee users and all corporate devices that are capable of 802.1X authentication use the employee
SSID. In the example network, an employee user has full access to all the network resources and the
internet. Guests use the guest SSID. Guest users are permitted to access the Internet using only
specific protocols such as HTTP and HTTPS. Only corporate devices that are not capable of 802.1X
authentication associate to the application SSID. These legacy devices that are not capable of 802.1X




are allowed to access only the necessary application servers. For example, a VoIP phone running SIP
can access only the SIP server to make calls. To optimize your network for multicast video
applications, see the Aruba Video Quick Reference & Design Guide.
File

Internet

Local
controller

File

Web

Internet

Local
controller

File

Web

Internet

PBX
RADIUS

Web
PBX

PBX
RADIUS

Data center

Local
controller

Data center

Guest
VLAN

Employee
VLAN

RADIUS

Data center

Application
VLAN

Employee
SSID
Application
SSID

Guest SSID
Employee
SSID

Figure 6

Application
SSID

Guest
SSID

Employee
SSID

Application SSID

arun_0361

Guest
SSID

Role-based access

The deployment scenario in this VRD portrays the needs of most campus deployments. However, the
requirements of each organization are different. Your network may differ from the VRD example
network in these ways:













VLAN and IP parameters
user density and VLAN pools
availability, redundancy, and performance requirements
type of devices on the network
applications running on the network
user role requirements
authentication and encryption requirements
SSID requirements
quality of service (QoS) requirements
intrusion detection and intrusion prevention requirements
mobility requirements
network management requirements

The network parameters and Aruba configurations shown in this VRD should be adjusted to meet your
needs.

Recommendations for Key Components
Some key components of this reference model include:
 Master controllers: The MMC-3600 controller is recommended for master controllers that do
not terminate any APs or AMs. The master controllers should be deployed in pairs for
redundancy. The master controller should be given adequate bandwidth connections to the
network, preferably a minimum of a Gigabit Ethernet LAN connection. A general best practice is




to configure each MMC-3600 controller in a full mesh with redundant links to separate data
center distribution switches. The MMC-3600 does not have redundant power supplies, so it is
recommended that you connect each appliance to discrete power sources in the data center.
Master
standby

arun_0410

Master
active

Figure 7


Master controllers deployed in full mesh topology

Local controllers: Use the M3 Controller Module for local controllers. In a pair of M3 controller
modules configured for local controller redundancy, each controller should have its own MMC6000 chassis. So two MMC-6000 chassis can accommodate four pairs of redundant local
controllers. Connect each blade to its own distribution layer switch with two 10 Gigabit Ethernet
connections with link aggregation. For configuration of link aggregation, see Appendix A: Link
Aggregation on page 149. The MMC-6000 chassis should contain redundant power supplies
connected to discrete power sources. See the mobility controller product line matrix to choose the
most appropriate mobility controller for your deployment.

!
CAUTION

Controllers that are redundant should not be placed in the same chassis,
because a chassis failure will cause the redundancy architecture to fail.

Two 10 Gigabit Ethernet links

Distribution

Distribution layer
switch

Local
mobility
controller

arun_052

Figure 8




Link Aggregation

Access points: Aruba offers a wide range of 802.11n APs. The spectrum capability and the
capacity of the APs vary. See the Aruba AP product line matrix to choose the most appropriate
AP for your deployment.
Air monitors: Deploy AMs at a ratio of approximately one AM for every four APs deployed, and
around the building perimeter for increased security and location accuracy. AMs perform many





of the intrusion detection system (IDS) duties for the network, including rogue AP containment.
AMs help to form accurate heat maps that display graphical RF data. Aruba considers dedicated
AMs to be a best practice for security because they provide full-time surveillance of the air. Use
the AP-105 as AMs, because these are dual-radio APs with full spectrum analysis support. For
details on the spectrum capabilities of all the Aruba APs, see the Aruba AP product line matrix.





Chapter 3: Master/Local Operation
Large campus deployments normally involve more than two controllers. When you have more than a
single pair of controllers, change control and network consistency can become an issue. To solve this
management scalability issue, Aruba Mobility Controllers can be deployed in clusters that consist of a
master and one or more local controllers. This design is the recommended model when two or more
controllers exist in the same network. This design is depicted in this VRD.
In an Aruba network that uses a master/local design, configuration is performed only on the master
and it is pushed down to the locals.

NOTE

In a large campus WLAN that has separate network services and aggregation
layers, APs and AMs should never terminate on the master controller. APs and
AMs should terminate only on the local controller. With this configuration, if the
master becomes unreachable or unavailable and no standby master has been
configured, the network continues to operate as expected, except for certain
operations. You cannot perform configuration, RF visualization, or location
services until connection to the master controller is restored. The master
controller is needed to perform configuration and reporting, but it is not a single
point of failure in the network.

Local controllers reside at the aggregation layer of the Aruba overlay architecture. They handle AP
termination, user authentication, and policy enforcement. When you configure any local controller, you
must know the IP address of the master and the pre-shared key (PSK) that was used to encrypt
communication between the controllers. The control channel between all Aruba controllers is protected
by an IP Security (IPsec) connection. For more details on the functions and responsibilities of master
and local mobility controllers in Aruba architecture, see the Aruba Mobility Controllers and Deployment
Models Validated Reference Design.

NOTE

The controllers have a preconfigured key at first boot. Change this key after the
first boot so that the operation of the master/local cluster is secure.

Controller Licenses
The ArubaOS™ base operating system contains many features and extensive functionality for the
enterprise WLAN network. Aruba uses a licensing mechanism to enable the additional features and to
enable AP capacity on controllers. The controller licensing depends on the user density and the
features needed to operate and secure your network. For more details about Aruba licenses, see
Aruba Mobility Controllers and Deployment Models Validated Reference Design.


Master/Local Operation | 19



Licensing Master Mobility Controllers
The master mobility controller must manage the functionality for all other platforms, so the master must
have the same license types as the local mobility controllers. Licensing unlocks the configuration
capabilities on the system. However, the master does not terminate APs or devices, so the master can
be licensed at a much lower level than the local mobility controller.
Only the functionality that is being enabled needs to be licensed. For example,
xSec is deployed primarily only in Federal Government and military
installations, and it is not required unless it will be in use at the organization.

NOTE

Table 2 lists the licenses that are used by the active and the standby master controllers in the example
network.
Table 2

Master Controller Licensing in the Example Network

License

Capacity

AP Capacity

0

PEF-NG

1

RFProtect

1

Licensing Local Mobility Controllers
Local controllers must be licensed according to the number of devices that consume licenses. Mobility
controllers should be licensed at the maximum expected capacity for that mobility controller. For
instance, in a failover scenario, the backup controller must be licensed to accept all the APs that it
could potentially host if a failure occurs, even if that is not the normal operating level.
In the example network, the two local mobility controllers are designed for active-active redundancy.
Each terminates a 40% load of APs and acts as the backup for the APs on the other controller. Each
controller is licensed to 80% of maximum capacity. If one mobility controller fails, the other controller
can add the additional APs from the failed controller.
Table 3 lists the licenses used by the local controllers in the example network.
Table 3

Local Controller Licensing in the Example Network

License

Capacity

AP Capacity

416

PEF-NG

416

RFProtect

416

Certificates
The Aruba controller comes with a default server certificate. This certificate demonstrates the secure
login process of the controller for captive portal, secure shell (SSH), and WebUI management access.
This certificate is not recommended for use in a production network. Aruba strongly recommends that




you replace this certificate with a unique certificate that is issued to the organization or its domain by a
trusted certificate authority (CA).
To receive a custom certificate from a trusted CA, generate a Certificate Signing Request (CSR) on
the controller and submit it to the CA. After you receive the digitally signed certificate from the CA,
import it to the controller. For more details about generating the CSR and importing certificates, see
“Managing Certificates” in the ArubaOS 6.1 User Guide available on the Aruba support site.





Chapter 4: VLAN Design and Recommendations
On an Aruba controller at the aggregation layer, VLANs are used in two logically different places:
 the access side of the controller where the APs terminate their GRE tunnels
 the user access side
VLANs are used on the access side of the controller where the APs terminate their GRE tunnels.
These VLANs carry traffic back and forth between APs and the controllers. Aruba strongly
recommends that edge access VLANs should not be dedicated to APs. The only exception where the
APs may have to be deployed on dedicated VLANs is in environments where 802.1X is a requirement
on the wired edge. The APs should use the existing edge VLANs as long as they have the ability to
reach the mobility controller. Deploying the APs and AMs in the existing VLANs allows for the full use
of the Aruba rogue detection capabilities. In pre 6.1 ArubaOS, the AMs had to be connected to a trunk
port that contains all VLANs that appear on any wired access port within range of the AM. This
connection was required for the AM to do wireless-to-wired correlation when tracking rogue APs. In
ArubaOS 6.1, the network administrators have the option of trunking all the VLANs available in the
access layer to the controller instead of trunking them to APs or AMs. Remember that all the access
VLANs should be trunked to every controller in the network that terminates APs and AMs. When all the
access VLANs are trunked to the controller, the controller assists the APs and AMs in wireless-towired correlation during rouge detection. Depending on your network design, you must choose
between trunking the VLANs to the controller or to the APs and AMs.

!
CAUTION

Wired containment requires that the hearing AP or AM is on the same subnet
as the contained device. If your access network has many VLANs and if you
want wired containment on all those VLANs, deploy the AMs on trunk ports.

Distribution-SW-1
145
145

145
Access switch

Figure 9


AP-LC1

arun_0356

LC1-Sunnyvale-6000
controller

AP plugged into a local switch, accessing the mobility controller

VLAN Design and Recommendations | 23



VLANs are also used on the user access side. On the user access side, user VLANs exist and traffic
flows to and from the users. During authentication, a process that is called “role derivation” assigns the
proper VLAN to each user and forwards traffic to the wired network if allowed. For campus networks,
Aruba recommends that you do not deploy the controllers as the default gateway for user VLANs. The
existing Layer 3 switches should remain the default gateways for all user VLANs. The Aruba
controllers should be deployed as a Layer 2 switched solution that extends from the distribution layer.
The controllers should be the default gateway and DHCP server only for the guest VLAN. For more
details about VLAN design, see the Aruba Mobility Controllers and Deployment Models Validated
Reference Design.

Distribution-SW-1

LC1-Sunnyvale-6000
controller

AP-LC1
150

Figure 10


arun_0358

Access switch

User VLAN, logical connection




VLAN Pooling
The Aruba VLAN pooling feature allows a set of VLANs to be assigned to a designated group of users.
VLAN pooling is tied to the virtual access point (VAP). Each VAP on a physical AP can have different
VLANs or VLAN pools. Aruba recommends using VLAN pools any time two or more user VLANs are
needed to support the user load from a single set of APs going to a single mobility controller. For more
details about VLAN pooling, see the Aruba Mobility Controllers and Deployment Models Validated
Reference Design.

VLANs 150, 151, 152, 153, 154
arun_049

Figure 11


arun_0359

Mobility
controller

VLAN pools distribute users across VLANs




To determine which pool to put the user into, the user MAC address is run through a hash algorithm.
The output of this algorithm places the user into one of the VLANs in the pool and ensures that the
user is always placed into the same pool during a roaming event. As the user associates with the next
AP, the address is hashed. The user is again placed into the same VLAN on the new AP, because the
hash algorithm generates the same output as before. The user can continue to use their existing IP
address with no break in their user sessions.

NOTE

The hashing algorithm does not place users into the available pool of VLANs in
a round-robin method. Ten clients that join a WLAN are not load balanced
equally among the VLANs. Instead, the distribution is based on the output of
the hash. One VLAN might have more users than the others. For example,
consider 150 clients that join a WLAN with just two VLANs in the pool and with
80 addresses per VLAN available for clients. Based on the output of the
hashing algorithm, 80 clients are placed in one VLAN and 70 in the other.
When the 151st client joins, the output of the hash might place the client in the
VLAN whose scope of 80 addresses has already exhausted. The result is that
the client cannot obtain an IP. To avoid such a rare situation, the network
administrator should design pools with sufficient number of user VLANs and
DHCP scopes to accommodate the user density.

A single VLAN or a VLAN pool can be named by the administrator. The VLAN names are global, but
the VLAN IDs associated with those names are local to the controller. The VLAN names are
configured globally in the master controller and are synchronized to the local controllers. The VLAN
IDs that are associated to a particular VLAN name are defined in the local controllers and can vary
across the controllers.

!
CAUTION

During VLAN pooling, the controller places the user into a particular VLAN
based on the hash calculated using the media access control (MAC) address of
the client. Hence, the VLAN obtained as a result of the hashing algorithm
cannot be predicted beforehand. In networks that use VLAN pooling, the clients
with static IP addressing will not work because the statically assigned VLAN
and the VLAN obtained by the controller after running the hash can be
different. Aruba recommends that VLAN pooling and static IP addressing
should never be used simultaneously within a single SSID.

The example network uses 10 VLANs (VLAN 150-159) split into these two pools:
 pool-7 is used by the employee and application VAPs in the AP group that uses the virtual IP
(VIP) of Virtual Router Redundancy Protocol (VRRP) instance 7 as the local management switch
(LMS) IP.
 pool-8 is used by the employee and application VAPs in the AP group that uses the VIP of
VRRP instance 8 as the LMS IP.





Table 4 lists the VLAN pools that are used in the example network.
Table 4

VLAN Pools in the Example Network

Pool Name

VLANs

pool-7

150-154

pool-8

155-158

CLI Configuration
MC1-Sunnyvale-3600
!
vlan-name pool-7 pool
!

LC1-Sunnyvale-6000
!
vlan pool-7 150-154
vlan pool-8 155-159
!

LC2-Sunnyvale-6000
!
vlan pool-7 150-154
vlan pool-8 155-159
!

WebUI Screenshot

Figure 12


VLAN pool on MC1-Sunnyvale-3600




Figure 13

Figure 14


VLAN pool on LC1-Sunnyvale-6000

VLAN pool on LC2-Sunnyvale-6000




Chapter 5: Redundancy
Aruba offers several redundancy models for master controller redundancy and local control
redundancy. The Aruba redundancy solutions can be implemented using VRRP or backup LMS IP.
Use VRRP, which operates at Layer 2, for redundancy whenever possible. For more details about the
various redundancy models and when to use backup LMS IP, see Aruba Mobility Controllers and
Deployment Models Validated Reference Design.

Master Redundancy
To achieve high availability of the master controller, use the master redundancy model. In this
scenario, two controllers are used at the network services layer: one controller is configured as the
active master and the other controller acts as standby master. This setup is known as “hot standby”
redundancy. The two controllers run a VRRP instance between them and the database and RF
planning diagram is synchronized periodically. The VIP address that is configured in the VRRP
instance is used by local mobility controllers, wired APs, and wireless APs that attempt to discover a
mobility controller. That VIP address is also used for network administration. The DNS query made by
APs to find the master controller resolves to this VIP. The synchronization period is a configurable
parameter with a recommended setting of 30 minutes between synchronizations.
Periodic database
synchronization

Master
active

Master
standby

arun_0226

VRRP keepalives

Figure 15

Hot-standby redundancy

In this configuration, one controller is always the active master controller and the other is always the
standby master controller. When the active controller fails, the standby controller becomes the active
master. Disable preemption in this setup. When preemption is disabled, the original master controller
does not automatically become the active master after it has recovered and instead acts as the backup
master controller. The recommended network attachment method is to have each master controller
configured in a full mesh with redundant links to separate data center distribution switches. The
example network uses a VRRP instance named 130 for redundancy.


Redundancy | 29



Table 5 and Table 6 summarize the VRRP instance used for master redundancy in the example
network.
Table 5

VRRP Instance Used for Master Redundancy

VRRP
ID

VRRP IP

Active Controller

Standby Controller

130

10.169.130.8

MC1-sunnyvale-3600
(Priority 110)

MC2-sunnyvale-3600
(Priority 100)

Table 6

Tracking
Master Up
Time

Tracking
Master Up
Time
Priority

30

20

Database Synchronization Parameters

Enable Periodic
Database Synchronization

Database Synchronization
Period in Minutes

Include RF Plan Data

enabled

30

enabled

CLI Configuration
MC1-Sunnyvale-3600
!
master-redundancy
master-vrrp 130
peer-ip-address 10.169.130.7 ipsec **********
!
vrrp 130
priority 110
ip address 10.169.130.8
description "Preferred-Master"
vlan 130
tracking master-up-time 30 add 20
no shutdown
!
database synchronize period 30
database synchronize rf-plan-data
!


Redundancy | 30



MC2-Sunnyvale-3600
!
master-redundancy
master-vrrp 130
peer-ip-address 10.169.130.6 ipsec **********
!
vrrp 130
ip address 10.169.130.8
description "Standby-Master"
vlan 130
tracking master-up-time 30 add 20
no shutdown
!
database synchronize period 30
database synchronize rf-plan-data
!

WebUI Screenshot

Figure 16


VRRP-130 on MC1-Sunnyvale-3600

Redundancy | 31



Figure 17

Figure 18


VRRP table on MC1-Sunnyvale-3600

VRRP-130 on MC2-Sunnyvale-3600

Redundancy | 32



Figure 19

VRRP table on MC2-Sunnyvale-3600

Local Redundancy
The local controllers at the aggregation layer also use VRRP instances to provide redundancy.
However, a different redundancy model called active-active redundancy is used. In this model, the two
local controllers terminate APs on two separate VRRP VIP addresses. Each Aruba controller is the
active local controller for one VIP address and the standby local controller for the other VIP. The
controllers share a set of APs and divide the load among them. The APs are configured in two different
AP groups, each with a different VIP as the LMS IP address.
Keepalives

Local

Local

Active VIP
Standby VIP

Active VIP
Standby VIP

Air monitor

Figure 20


arun_0264

arun_044

Active-active redundancy

Redundancy | 33



When an active local controller becomes unreachable, the APs that are managed by that controller fail
over to the standby controller for that VRRP instance. Under these conditions, one controller
terminates the entire AP load in the network. Therefore each controller must have sufficient processing
power and licenses to accommodate all of the APs that are served by the entire cluster. Though the
controllers are designed to support 100% capacity, do not load the mobility controllers past the 80%
capacity so that the network is more predictable and allows headroom. Aruba recommends that each
mobility controller be run at only 40% capacity, so that when a failover occurs, the surviving mobility
controller carries only an 80% load. This load gives the mobility controller the room to operate under
the failover conditions for a longer period of time.
In this model, preemption should be disabled so that APs are not automatically forced to fail back to
the original primary controller after it recovers. Whenever an AP fails over to a different controller, all
the clients served by that AP get disconnected. So if a controller malfunctions and reboots constantly,
then the APs served by that controller will “flap” between the original controller and standby controller if
preemption is enabled. When preemption is disabled, the network administrator has sufficient time to
troubleshoot the issue without this ping pong effect. The APs do not automatically fail back to the
original controller, so this model requires that the mobility controller is sized appropriately to carry the
entire planned failover AP capacity for an extended period of time.
Table 7 summarizes the VRRP instances used for local redundancy in the example network.
Table 7

VRRP Instances Used for Local Redundancy

VRRP ID

VRRP IP

Active Controller

Standby Controller

7

10.169.145.7

LC1-Sunnyvale-6000 (Priority 110)


8

10.169.145.8



CLI Configuration
LC1-Sunnyvale-6000
!
vrrp 7
priority 110
ip address 10.169.145.7
description "intial-primary-7"
vlan 145
no shutdown
!
vrrp 8
ip address 10.169.145.8
description "initial-standby-8"
vlan 145
no shutdown
!


Redundancy | 34



LC2-Sunnyvale-6000
!
vrrp 7
ip address 10.169.145.7
description "initial-standby-7"
vlan 145
no shutdown
!
vrrp 8
priority 110
ip address 10.169.145.8
description "initial-primary-8"
vlan 145
no shutdown
!

WebUI Screenshot

Figure 21


VRRP-7 on LC1-Sunnyvale-6000

Redundancy | 35



Figure 22

Figure 23



VRRP table on LC1-Sunnyvale-6000

Redundancy | 36



Figure 24

Figure 25




Redundancy | 37



Figure 26


VRRP table on LC2-Sunnyvale-6000

Redundancy | 38



Chapter 6: User Roles, Profiles, and AP Groups
In the Aruba user-centric network, every client is associated with a user role. The user roles that are
enforced through the firewall policies determine the network privileges of a user. A policy is a set of
rules that applies to the traffic that passes through the Aruba devices. The rules and policies are
processed in a top-down fashion, so the position of a rule within a policy and the position of a policy
within a role determine the functionality of the user role. When you construct a role, you must put the
rules and policies in the proper order.
The PEFNG license is essential to exploit the identity-based security features on the Aruba controller.
The PEFNG license also adds a set of predefined policies on the controller, which can be used or
modified as required.

!
CAUTION

Modifying the predefined policies is not recommended. If necessary, create a
new policy that depicts the predefined rule and then customize it.

The type of user roles and polices vary between organizations and the example network defines roles
and policies that are implemented in most cases. In the example network, the following roles are used:
 employee role
 application role
 guest-logon role
 auth-guest role
Figure 27 summarizes the user roles used in the example network and all the policies associated with
each of those roles.
User roles
• employee
• application
• guest-logon
• auth-guest

Firewall policies

Firewall policies

Firewall policies

• common-policy

• sip-session-allow

• captiveportal (predefined policy)

• cplogout (predefined policy)

• sip-session-allow

• dhcp-acl (predefined)

• guest-logon-access

• guest-logon-access

• ocs-lync

• tftp-session-allow

• block-internal-access

• block-internal-access

• allowall (predefined)

• dns-acl (predefined)

• auth-guest-access

• icmp-acl (predefined)

• drop-and-log

Figure 27


arun_0367

Firewall policies

User roles used in the example network

User Roles, Profiles, and AP Groups | 39



Alias
The Alias feature in the ArubaOS can be used to group several hosts or networks. Use this option
when several rules have protocols and actions common to multiple hosts or networks. An alias
simplifies a firewall policy by reducing the number of ACL entries. The alias allows IP addresses to be
added by host, network, or range. When the invert parameter of an alias is enabled, the rules that use
that alias are applied to all the IP addresses except those specified in the alias. For more information
about alias, see the ArubaOS 6.1 User Guide available on the Aruba support site.
Table 8 lists the aliases that are used in the example network.
Table 8

Aliases

Alias Name

Purpose

IP Address/ Range

Public-DNS

Defines the public DNS servers

Host
8.8.8.8
216.87.84.209

Internal-Network

Defines the private IPv4 address range

Network
10.0.0.0/8
172.16.0.0/16
192.168.0.0/24

sip-server

Defines the SIP servers in the network

Host
10.169.130.20

tftp-server

Defines the TFTP servers in the network

Host
10.169.130.11

dns-servers

Defines the internal DNS servers

Host
10.169.130.4

ocs-lync

Defines the Microsoft Lync servers

Host
10.169.130.35

Amigopod

Defines the Amigopod server

Host
10.169.130.50





Configuration Profiles
Configuration profiles allow different aspects of the Aruba WLAN to be grouped into different
configuration sets. Each profile is essentially a partial configuration. SSID profiles, radio profiles, and
AAA profiles are just some of the available choices. For more information about these profiles, see the
Aruba 802.11n Networks Validated Reference Design and ArubaOS 6.1 User Guide.
Figure 28 shows an overview of the profile structure and high-level overview of an AP group.
WLAN

AP group
Virtual AP profile

AAA

AAA profile

AP name
(individual AP)

MAC auth profile

User roles

QOS

VoIP CAC

User derivation
rule profile

Traffic management
802.1X auth profile

RF

802.11a radio
profile
802.11b/g radio
profile

802.1X auth
server group

ARM profile

MAC auth
server group

Authentication
servers

HT radio profile
RF event
thresholds

Spectrum profile

RF optimization

AM scanning
profile

SSID

EDCA station
profile

SSID profile
EDCA AP
profile
High throughput
profile

AP

Wired AP profile
Regulatory domain
profile
AP system
profile

WMM traffic management
Ethernet0
profile

Ethernet1
profile
SNMP profile

802.11k
SNMP user
profile
IDS general

Mesh radio
profile

IDS signature
matching

IDS signature
profiles

IDS DoS profile

IDS

IDS rate
thresholds

IDS profile

Mesh cluster
profile

Figure 28


Mesh high-throughput
SSID profile

Mesh

IDS impersonation
profile
IDS unauthorized
device profile

arun_0344

Mesh radio
profile

High-level overview of an AP group




AP Groups
An AP group is a unique combination of configuration profiles. In general, all profiles can be assigned
to an AP group to create a complete configuration. This flexibility in configuration allows arbitrary
groupings of APs such as “All Headquarter APs”, “All Lobby APs”, or “All AMs”, with different
configurations for each. Configuration profiles provide flexibility and convenience to wireless network
managers who create AP groups. An AP group must include a minimum number of profiles, in
particular, a VAP profile.
Each AP, AM, SM, and RAP can be a part of only one AP group at any one
time. This limitation eliminates the need to merge possibly contradictory
configurations and prevents multiple VAPs with the same SSID from being
enabled on the same physical AP.

NOTE

The example network uses the following four AP groups:
 AP-LC1-Sunnyvale-6000
 AM-LC1-Sunnyvale-6000
 AP-LC2-Sunnyvale-6000
 AM-LC2-Sunnyvale-6000
Figure 29 summarizes the configuration profiles used by these four AP groups in the example network.
The chapters that follow explain how to configure each of these profiles and why they are necessary.
SSID profiles

AAA profiles

VAP profiles

• corp-employee

• corp-employee

• corp-employee-LC1-Sunnyvale-6000

• corp-app

• corp-app

• corp-app-LC1-Sunnyvale-6000

• guestnet

• guestnet

• guestnet

AP system profiles

ARM profile

AM scanning profiles

802.11a radio profiles

• LC1-Sunnyvale-6000

• corp-arm

• am-scan

• airmonitor

• LC2-Sunnyvale-6000

• AP

Traffic management profiles

VoIP call admission

IDS profile

• airmonitor

• traffic-LC1-Sunnyvale-6000

control profile

• corp-wips

• AP

• traffic-LC2-Sunnyvale-6000

• corp-voip-cac

Figure 29


arun_0364

802.11g radio profiles

All the profiles configured in the example network




Chapter 7: AP Groups for Client Access
When you create an AP group for client access you create a functional WLAN for client access. To
create an AP group for client access, you need to configure these:
 firewall policies and user roles (required)
 SSID profile (required)
 server groups, AAA profile (required)
 VAP profile (required)
 Adaptive Radio Management (ARM) profile (optional, but recommended)
 802.11a radio profile (required)
 802.11g radio profile (required)






AP system profile (required)
802.11a traffic management profile (optional, but recommended)
802.11g traffic management profile (optional, but recommended)
VoIP call admission control profile (optional, but recommended)
IDS profile (optional, but recommended)

The following chapters explain the configuration of the AP-LC1-Sunnyvale-6000 and AP-LC2Sunnyvale-6000 AP groups for client access. The AP-LC1-Sunnyvale-6000 AP group is used for all
APs that must be managed by the LC1-Sunnyvale-6000 local controller. The AP-LC2-Sunnyvale-6000
AP group is used for all APs that must be managed by the LC2-Sunnyvale-6000 local controller.
The APs that belong to one of these two AP groups perform these actions:
 Broadcast employee, application, and guest SSIDs in both 2.4 GHz and 5 GHz bands.
 Participate in ARM and band steering.
 Terminate on VRRP-7 VIP if they are in the AP-LC1-Sunnyvale-6000 AP group or on VRRP-8
VIP if they belong to the AP-LC2-Sunnyvale-6000 AP group.




Participate in prioritizing traffic based on SSID and WMM category.
Participate in VoIP call admission control.
Participate in wireless intrusion prevention.


AP Groups for Client Access | 43



Figure 30 summarizes the profiles used for the AP-LC1-Sunnyvale-6000 and AP-LC2-Sunnyvale-6000
AP groups.
AP groups for client access
• AP-LC1-Sunnyvale-6000
• AP-LC2-Sunnyvale-6000

AP group =
AP-LC1-Sunnyvale-6000

AP group =
VAP = • corp-employee-LC2-Sunnyvale-6000
• guestnet

802.11a radio profile = AP

802.11a radio profile = AP

802.11g radio profile = AP

802.11g radio profile = AP

AP system profile = LC1-Sunnyvale-6000

AP system profile = LC2-Sunnyvale-6000

802.11a traffic management profile = traffic-LC1-Sunnyvale-6000

802.11a traffic management profile = traffic-LC2-Sunnyvale-6000

802.11g traffic management profile = traffic-LC1-Sunnyvale-6000

802.11g traffic management profile = traffic-LC2-Sunnyvale-6000

VoIP call admission control profile = corp-voip-cac

VoIP call admission control profile = corp-voip-cac

IDS profile = corp-wips

IDS profile = corp-wips

Figure 30


arun_0385

VAP = • corp-employee-LC1-Sunnyvale-6000
• guestnet

AP groups for client access

AP Groups for Client Access | 44



Chapter 8: Configuring the Employee Role
Company employees can be granted a role based on their specific job function, department, domain,
or they can be given a universal employee role. In certain organizations, users typically are placed in a
single user role that has access to all internal and external resources.
The employee role used in the example network grants unrestricted access to the internal and external
resources, but denies personal DHCP servers. The employee role is the default role assigned to
clients after successful 802.1X authentication to the employee SSID. This role gives voice traffic
priority over standard data traffic. All company employees and devices capable of 802.1X/EAP use the
employee SSID.
Before you configure the employee role, first you must create the policies associated with it. The
employee role in the example network uses the following policies:
 common
 sip-session-allow
 ocs-lync
 allowall (predefined)

Configuring the Common Policy
In most campus deployments, all users should be denied certain services, no matter what their roles.
This common policy is used by the employee role to do the following things:
 Deny users from activating their personal DHCP servers on the network.
 Allow ping across the network.
 Allow DNS queries only to certain predefined DNS servers.
Remember, the order of rules within a policy determines the behavior of the policy.


Configuring the Employee Role | 45



Table 9 summarizes the rules used for the common policy.
Table 9

Rules Used for the Common Policy

Rule
Source
Number

Destination

Service

Action

Purpose

1

User

Any

UDP
min port = 68
max port = 68

Drop

This rule drops responses from a personal
DHCP server, which prevents the clients
from acting as DHCP servers.

2

Any

Any

Service
svc-dhcp (udp 67 68)

Permit

This rule allows clients to request and
discover a DHCP IP address over the
network. The DHCP server on the network
does not fall under the User category, so its
response on port 68 is not dropped by the
first rule. The first two rules guarantee that
DHCP is processed only by legitimate DHCP
servers on the network.

3

Any

Any

Service
svc-icmp (icmp 0)

Permit

This rule allows ICMP (ping) across the
internal network.

4

Any

Alias
dns-servers

Service
svc-dns (udp 53)

Permit

This rule allows DNS queries to the set of
DNS servers defined in the dns-servers
alias.

CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session common
user any udp 68 deny
any any svc-dhcp permit
any any svc-icmp permit
user alias dns-servers svc-dns permit
!





WebUI Screenshot
MC1-Sunnyvale-3600

Figure 31

Common policy

Configuring the sip-session-allow Policy
The sip-session-allow policy prioritizes SIP traffic and allows SIP services only between the user and
the corporate PBX and servers that provide voice service. If the organization supports protocols such
as NOE from Alcatel Lucent, H.323, SCCP, Vocera, and others for voice communication, policies
should be created to prioritize them.
Table 10 summarizes the rules used for the sip-session-allow policy.
Table 10

Rules Used for the sip-session-allow Policy

Rule
Source
Number

Destination

Service

Action

Queue

Purpose

1

user

Alias sip-server

Service
svc-sip-udp

permit

high

Allows SIP sessions between users
and SIP servers using the UDP
protocol

2

user

Alias sip-server

Service
svc-sip-tcp

permit

high

Allows SIP sessions between users
and SIP servers using the TCP
protocol

3

Alias
sip-server

user

Service
svc-sip-udp

permit

high

Allows SIP sessions between SIP
servers and users using the UDP
protocol

4

Alias
sip-server

user

Service
svc-sip-tcp

permit

high

Allows SIP sessions between SIP
servers and users using the TCP
protocol





CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session sip-session-allow
user alias sip-server svc-sip-udp permit queue high
user alias sip-server svc-sip-tcp permit queue high
alias sip-server user svc-sip-udp permit queue high
alias sip-server user svc-sip-tcp permit queue high
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 32

sip-session-allow policy

Configuring the ocs-lync Policy
Many organizations use Microsoft Office Communications Server or Microsoft Lync Server for voice,
instant messaging, and conferencing. These Microsoft products use SIP over TLS for signaling.
ArubaOS performs a deep packet inspection on traffic established over a secure channel to identify
the voice and video sessions. This deep packet inspection of encrypted traffic allows Aruba to provide
QoS for the voice or video sessions established even over the secure layers such as TLS or IP Sec.
The classify media option enables this support.
Microsoft OCS/Lync uses TCP on port 5060/5061 for communication between the Microsoft OCS/Lync
server and Office Communicator /Lync client. So, the ocs-lync policy is constructed to perform deep
packet inspection only on the TCP traffic on ports 5060/5061.





Table 11 summarizes the rules used for the ocs-lync policy.
Table 11

Rules Used for the ocs-lync Policy

Rule
Number

Source

Destination Service

Action

Queue

Classify
Media

1

User

Alias
ocs-lync

Service
svc-sips (tcp
5061)

permit

high

Enabled

2

User

Alias
ocs-lync

Service
svc-sip-tcp (tcp
5060)

Permit

high

Enabled

3

Alias
ocs-lync

any

Service
svc-sips (tcp
5061)

Permit

high

Enabled

4

Alias
ocs-lync

any

Service
svc-sip-tcp (tcp
5060)

permit

high

Enabled

!
CAUTION

Purpose
This rule performs deep
packet inspection and
prioritization of encrypted
voice, and video sessions
of OCS/Lync.

Selecting any for the service field and setting the classify media flag, has an
impact on performance because it turns on deep packet inspection for all traffic
types. Choose this option only for services that require it.

CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session ocs-lync
user alias ocs-lync svc-sips permit classify-media queue high
user alias ocs-lync svc-sip-tcp permit classify-media queue high
alias ocs-lync any svc-sips permit classify-media queue high
alias ocs-lync any svc-sip-tcp permit classify-media queue high
!





WebUI Screenshot
MC1-Sunnyvale-3600

Figure 33

ocs-lync policy

After all the required policies are configured, place the required firewall policies in correct order to
create the employee role. Remember, the order of policies determines the behavior of a user role.
Table 12 summarizes the policies in the employee role.
Table 12

Policies in the Employee Role

Policy
Number

Policy Name

Purpose

1

common

This policy denies personal DHCP servers but allows legitimate DHCP and DNS
services. For details, see Configuring the Common Policy on page 45.

2

sip-session-allow

This policy gives voice traffic priority using the high priority queue. For details, see
Configuring the sip-session-allow Policy on page 47.

3

ocs-lync

This policy gives priority to encrypted voice and video sessions used by Microsoft OCS
and Lync services. For details, see Configuring the ocs-lync Policy on page 48.

4

allowall (predefined)

This policy allows any service from any source to any destination.





CLI Configuration
MC1-Sunnyvale-3600
!
user-role employee
access-list session
access-list session
access-list session
access-list session
!

common
SIP-session-allow
ocs-lync
allowall

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 34


Employee role




Chapter 9: Employee VAP Profiles
A typical home AP advertises only one SSID, so even with a dual-radio AP, only two WLANs can be
formed. Ideally, in these situations the number of physical APs is proportional to the number of WLANs
supported. Aruba solves this issue with the concept of virtual access points (VAPs). VAPs are logical
entities that are present within a physical AP.
Physical Aruba APs, unlike typical home APs, are often configured to appear as more than one
physical AP. This configuration provides the necessary authentication and encryption combinations
without collocating excessive amounts of APs in the same physical area.
The VAPs share the same channel and power settings on the radio, but each appears as a separate
AP with its own SSID (ESSID) and MAC address (BSSID).

Guest
SSID

Employee
SSID
Application SSID

Figure 35

arun

A typical set of VAPs on one physical AP

Aruba supports up to eight BSSIDs per radio on the AP, with a maximum of 16 VAPs per physical AP.
The maximum total number of BSSIDs that are supported across the WLAN are a function of the
mobility controller model. For more on these limitations, see the mobility controller matrix and AP
matrix on the Aruba networks public site at http://www.arubanetworks.com/vrd.

!
CAUTION


Aruba does not recommend running an AP with the maximum number of VAPs
available. Each VAP acts like a real AP and is required to beacon like any other
AP. This beaconing consumes valuable airtime that would otherwise be used
by clients to transmit data on the channel. Aruba recommends that you
leverage the smaller numbers of SSIDs and user roles and deploy a new SSID
only when a new encryption or authentication type is required.

Employee VAP Profiles | 53



The BSSID assigned to each of the 16 possible SSIDs on a physical AP are
generated from the MAC address of the physical AP. A total of 16 BSSIDs are
generated by the algorithm. The BSSID assigned to each SSID is random.
Whenever an AP reboots the BSSID to SSID mapping may change. In certain
situations a SSID may be temporarily disabled for maintenance, when this
SSID is enabled again, the BSSID assigned to it might not be the same as
before.

NOTE

A VAP profile is a container that holds an AAA profile, SSID profile, 802.11k profile, and WMM traffic
management profile. At minimum, each VAP profile must have an AAA and SSID profile. The VAP
profile also has other configurable features, such as band steering, dynamic multicast optimization,
fast roaming, and DoS prevention.
Table 13 summarizes the VAP profiles used in the example network.
Table 13

VAP Profiles Used in the Example Network

VAP Profile

AP Group

VLAN/ VLAN Pool

Corp-Employee-LC1-Sunnyvale-6000


pool-7

Corp-App-LC1-Sunnyvale-6000


pool-7



pool-8



pool-8

guestnet

AP-LC1-Sunnyvale-6000,

900

Configuring the SSID Profiles
Service Set Identifier (SSID) is the network or WLAN that any client sees. A SSID profile defines
parameters, such as name of the network, authentication type for the network, basic rates, transmit
rates, SSID cloaking, and certain WMM settings for the network.
Aruba offers different flavors of the Advanced Encryption Standard (AES), Temporal Key Integrity
Protocol (TKIP), and wired equivalent privacy (WEP) encryption. AES is the most secure and
recommended encryption method. Most modern devices are AES capable and AES should be the
default encryption method. Use TKIP only when devices that are not AES capable are present. In
these situations, use a separate SSID for devices that are only capable of TKIP. It is important to
understand that several vulnerabilities have been reported with TKIP. Avoid using WEP, because it
can be cracked in less than 5 minutes with generally available tools. Aruba also supports a host of
authentication methods such as 802.1X, captive portal, PSK, and WEP.





Configuring the Employee SSID Profile
By default, all employees and corporate devices should connect to the employee SSID. The use of
802.1X with a backend RADIUS server and AES encryption makes this the most secure network. For
more information about authentication types, encryption methods, and role derivation on the Aruba
Mobility Controller for Wi-Fi Protected Access® 2 (WPA2™), see the Aruba 802.11n Networks
Validated Reference Design.
Configuring Wi-Fi Multimedia
Wi-Fi Multimedia™ (WMM®) is a Wi-Fi Alliance® certification program that is based on the IEEE
802.11e amendment. WMM ensures QoS for latency-sensitive traffic in the air. WMM divides the traffic
into four queues or access categories:
 voice
 video
 best effort
 background
The traffic is prioritized based on the queue it belongs to. The order of priority is voice > video > best
effort > background. Like WMM for QoS in air, QoS on the wired side of the network is dictated by the
DiffServ Code Point (DSCP) and 802.1p tagging. To ensure end-to-end QoS on the network, consider
these requirements:
 The DSCP tags should translate to appropriate WMM access categories and vice-versa. The
Aruba infrastructure ensures this translation between WMM and DSCP/802.1P markings when
the traffic moves across wired and wireless mediums.


All devices in the network should be capable of and configured for QoS support. The LAN that is
between the AP and the mobility controller must recognize and prioritize DCSP-marked traffic
through the network. Similarly, the core must respect the DSCP marks from the mobility
controller to the multimedia servers.

For more information about the mapping between WMM access categories, DSCP tags and other QoS
functionalities, see the Aruba 802.11n Networks Validated Reference Design.
In the example network, the WMM parameter is enabled on the employee SSID to prioritize latencysensitive traffic, such as voice and video, over the standard data traffic. The DSCP-to-WMM mapping
is a configurable parameter that is available within the SSID profile. In the example network, the
DSCP-to-WMM mapping values are set to the defaults. The Aruba default DSCP-to-WMM mapping
values match the default DSCP settings of most vendors. Alter the Aruba defaults only if they vary
from your existing DSCP settings.





Table 14 summarizes the Corp-Employee SSID profile.
Table 14

Corp-Employee SSID Profile

Network
SSID Profile Name
(SSID)

Authentication

Encryption

WMM

Purpose

Corp-Employee

WPA2

AES

Enabled

All employees and corporate
devices that support 802.1X will
be in this SSID.

Employee

CLI Configuration
MC1-Sunnyvale-3600
!
wlan ssid-profile "Corp-Employee"
essid "Corp-Employee"
opmode wpa2-aes
wmm
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 36


Corp-Employee SSID




Figure 37
WMM enabled for Corp-Employee SSID
(available on the Advanced tab of the SSID profile)

Configuring the AAA Profiles
The AAA profiles define how users are authenticated. The AAA profile determines the user role for
unauthenticated clients (initial role) and the user role to be applied after successful authentication
(default role) based on the authentication type. The AAA profile also defines the server group that is
used for the defined authentication method and RADIUS accounting. For example, consider two
different locations, Sunnyvale and New York, where the employee WLAN should be available and
each location has its own RADIUS server. The employee SSID profile is the same, but there will be
two AAA profiles: one for Sunnyvale and one for New York, because two different servers exist for
authentication. So, APs in Sunnyvale will have a different VAP for the employee WLAN than APs in
New York.
Authentication Server and Server Groups
For authentication, ArubaOS can use the internal database or external authentication servers such as
RADIUS, LDAP, TACAS+, and Windows server. A server group is a collection of servers used for
authentication. In case of 802.1X authentication, the external RADIUS server or servers used for
802.1X authentication for a particular WLAN are grouped together as a server group. By default, the
first server on the list is used for authentication unless it is unavailable. A server group can have
different authentication servers. For example, you can create a server group that uses an LDAP server
as a backup for a RADIUS server.





Configuring the NPS Server Group for 802.1X Authentication
The example network uses the server group named NPS for 802.1X authentication of corporate users.
A RADIUS server called NPS1 is defined and added to the NPS server group. For details on 802.1X/
EAP process, see the Aruba 802.11n Networks Validated Reference Design.

NOTE

If the RADIUS server is configured to return specific attributes for the users
after authentication, then the server-derived role that corresponds to the
returned attributes can be configured under server groups. For information
about configuring a server-derived role, see the ArubaOS 6.1 User Guide
available on the Aruba support site.

CLI Configuration
MC1-Sunnyvale-3600
!
aaa authentication-server radius "NPS1"
host "10.169.130.20"
key **********
timeout 30
!
aaa server-group "NPS"
auth-server NPS1
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 38


NPS1 RADIUS Server




Figure 39

NPS sever group

Configuring the Employee AAA Profile
A AAA profile named corp-employee is used for the employee WLAN. First create a AAA profile called
corp-employee and then configure the following parameters in it:
 Default role for 802.1X authentication: employee role (see Chapter 8: Configuring the Employee
Role on page 45).
 802.1X authentication server group: NPS
 802.1X profile:
 Create the corp-employee-dot1x 802.1X profile.
 Enable termination. (By default the Termination EAP-Type is eap-peap and Termination Inner
EAP Type is eap-mschapv2.)

NOTE

Aruba recommends 802.1X termination on the controller. This feature, also
known as AAA FastConnect™, offloads the cryptographic portion of 802.1X/
EAP authentication exchange to the controller, which reduces the load on the
RADIUS server. AAA FastConnect permits several hundred authentication
requests per second to be processed, which increases authentication server
scalability. This feature is very useful when the authentication server is not
802.1X capable, such as an LDAP server. For details about AAA FastConnect,
see the Aruba 802.11n Networks Validated Reference Design.

CLI Configuration
MC1-Sunnyvale-3600
!
aaa authentication dot1x "corp-employee-dot1x"
!
!
aaa profile "corp-employee"
authentication-dot1x "corp-employee-dot1x"
dot1x-default-role "employee"
dot1x-server-group "NPS"
!





WebUI Screenshot
MC1-Sunnyvale-3600

Figure 40

corp-employee-dot1x 802.1X authentication profile

Figure 41


corp-employee AAA profile




Configuring the Employee VAP Profiles
Band steering, which is a feature of ARM, identifies dual-band-capable clients and can encourage or
force them to move to the 5 GHz band. The 5 GHz band has more channels, more bandwidth
availability, and fewer sources of interference than the 2.4 GHz band. Aruba recommends using band
steering. Figure 42 summarizes the employee VAPs used in the example network.
Employee VAPs

vlan = pool-8
Band steering = prefer 5 GHz

SSID profile
• corp-employee

SSID profile
• corp-employee

AAA profile
• corp-employee

AAA profile
• corp-employee

Figure 42

arun_0366

vlan = pool-7

Employee VAP profiles

Table 15 lists the parameters that are configured for the Corp-Employee-LC1-Sunnyvale-6000 and
Corp-Employee-LC2-Sunnyvale-6000 VAP profiles.
Table 15

Employee VAP Profiles

VAP Profile

VLAN

Band Steering

AAA Profile

SSID Profile


pool-7

-Enabled
-Prefer 5 GHz

corp-employee

Corp-Employee


pool-8

-Enabled
-Prefer 5 GHz

corp-employee

Corp-Employee





CLI Configuration
MC1-Sunnyvale-3600
!
wlan virtual-ap "Corp-Employee-LC1-Sunnyvale-6000"
aaa-profile "corp-employee"
ssid-profile "Corp-Employee"
vlan pool-7
band-steering
wmm-traffic-management-profile "corp-wmm"
!
wlan virtual-ap "Corp-Employee-LC2-Sunnyvale-6000"
aaa-profile "corp-employee"
ssid-profile "Corp-Employee"
vlan pool-8
band-steering
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 43


Corp-Employee-LC1-Sunnyvale-6000 VAP profile



Figure 44



Corp-Employee-LC2-Sunnyvale-6000 VAP profile




Chapter 10: Configuring the Application Role and VAP Profiles
Certain devices, such as legacy handheld scanners and IP video cameras, are not capable of 802.1X/
EAP and use PSK for authentication. In cases where PSK has be to supported to accommodate the
devices that do not support 802.1X, you must create a separate user role for those applications and
devices. Unlike the employee role, this user role should be restricted only to the services and servers
required by such devices.
The example network has some SIP phones that are not capable of 802.1X and use the application
SSID. These phones use the SIP protocol and need TFTP to download configurations. These phones
in the example network need only SIP, DHCP, TFTP, DNS, and ICMP services to operate. Different
devices use different protocols for operation. The services that the devices require depend on the
vendor. Contact your device vendor to determine the services that are needed for your device
operation.
The example network assigns the application user role to the devices that associate to the application
SSID. The application role allows access only to the SIP, DHCP, TFTP, DNS, and ICMP services. The
application role ensures that the devices that associate to the application SSID access only the
required services and servers.
Before you configure the application role, first create the policies that are associated with it. The
application role in the example network uses the following policies:
 sip-session-allow (For details, see Configuring the sip-session-allow Policy on page 47.)
 dhcp-acl (predefined)
 tftp-session-allow
 dns-acl (predefined)
 icmp-acl (predefined)


Configuring the Application Role and VAP Profiles | 65



Configuring the tftp-session-allow Policy
The tftp-session-allow policy allows only TFTP services between the user and TFTP servers.
Table 16 summarizes the rules used for the tftp-session-allow policy.
Table 16

Rules Used for the tftp-session-allow Policy

Rule
Number

Source

Destination

Service

Action

Purpose

1

user

Alias tftp-server

Service
svc-tftp

permit

Allows TFTP sessions between the
user and TFTP servers.

CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session tftp-session-allow
user alias tftp-server svc-tftp permit
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 45


tftp-session-allow policy




Configuring the Application Role
To create the desired application role, you must order the essential firewall policies properly.
Table 17 summarizes the order of the policies in the application role that is used by the example
network.
Table 17

Policies in the Application Role

Policy
Number

Policy Name

Purpose

1

sip-session-allow

Allows SIP service. For details, see Configuring the sip-session-allow Policy on
page 47.

2

dhcp-acl (predefined)

Allows DHCP service.

3

tftp-session-allow

Allows TFTP service. For details, see Configuring the tftp-session-allow Policy on
page 66.

4

dns-acl (predefined)

Allows DNS service.

5

icmp-acl (predefined)

Allows ICMP across the network.

CLI Configuration
MC1-Sunnyvale-3600
!
user-role application
access-list session sip-session-allow
access-list session dhcp-acl
access-list session tftp-Session-allow
access-list session dns-acl
access-list session icmp-acl
!





WebUI Screenshot
MC1-Sunnyvale-3600

Figure 46

Application role

Configuring the Application SSID Profile
The application SSID should be used strictly for devices that are not 802.1X capable. The application
SSID uses WPA2-PSK for authentication and AES for encryption. PSKs are susceptible to social
engineering attacks and offline dictionary attacks. The passphrase and key that is used should be at
least 20 characters. To protect against social engineering attacks, the passphrase and key should not
be distributed to everyone. Only the network administrators should know the passphrase.
In the example network, the WMM parameter is enabled to prioritize latency-sensitive applications,
and the DSCP-to-WMM mapping values are set to defaults.
Table 18 summarizes the Corp-App SSID profile.
Table 18

Corp-App SSID Profile

SSID
Profile

Network
Authentication
Name (SSID)

Encryption

WMM

Purpose

Corp-App

Application

AES

Enabled

Only for legacy devices that are
not 802.1X capable.


WPA2-PSK




CLI Configuration
MC1-Sunnyvale-3600
!
wlan ssid-profile "Corp-App"
essid "Application"
opmode wpa2-psk-aes
wmm
wpa-passphrase **********
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 47


Corp-App SSID profile




Configuring the Application AAA Profile
A AAA profile named corp-app is used for the application WLAN. PSK is used for authentication, so
the default role that is assigned to authenticated users is specified in the initial role parameter of the
AAA profile. To reduce the number of profiles, Aruba has included the default-psk profile within the
802.1X profile. The profiles are combined because the dynamic key generation process of a WPA™/
WPA2 PSK process is similar to that key generation process of 802.1X/EAP. The PSK passphrase is
run through an algorithm that converts it into a pairwise master key (PMK). This PMK is used in the
four-way handshake process to generate the dynamic encryption keys. Select the predefined profile
named default-psk as the 802.1X profile when PSK is used for authentication.
The following parameters are configured in the corp-app AAA profile:
 Initial Role: application role (see Configuring the Application Role on page 67)
 802.1X Profile: default-psk (predefined)

NOTE

If you do not assign an 802.1X profile in the AAA profile that is used for PSK,
connectivity issues may occur.

CLI Configuration
MC1-Sunnyvale-3600
!
aaa profile "corp-app"
initial-role "application"
authentication-dot1x "default-psk"
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 48


corp-app AAA profile




Configuring the Application VAP Profiles
Figure 49 summarizes the application VAP profiles used in the example network.
Application VAPs

vlan = pool-8

SSID profile
• corp-app

SSID profile
• corp-app

AAA profile
• corp-app

AAA profile
• corp-app

Figure 49

arun_0386

vlan = pool-7

Application VAP profiles

Table 19 lists the parameters that are configured for the Corp-App-LC1-Sunnyvale-6000 and CorpAPP-LC2-Sunnyvale-6000 VAP profiles.
Table 19

Application VAP Profiles

VAP Profile

VLAN

Band Steering

AAA Profile

SSID Profile


pool-7

-Enabled
-Prefer 5 GHz

corp-app

Corp-App


pool-8

-Enabled
-Prefer 5 GHz

corp-app

Corp-App

CLI Configuration
MC1-Sunnyvale-3600
!
wlan virtual-ap "Corp-App-LC1-Sunnyvale-6000"
aaa-profile "corp-app"
ssid-profile "Corp-App"
vlan pool-7
band-steering
!
wlan virtual-ap "Corp-App-LC2-Sunnyvale-6000"
aaa-profile "corp-app"
ssid-profile "Corp-App"
vlan pool-8
band-steering
!





WebUI Screenshot
MC1-Sunnyvale-3600

Figure 50


Corp-App-LC1-Sunnyvale-6000 VAP profile



Figure 51



Corp-App-LC2-Sunnyvale-6000 VAP profile




Chapter 11: Configuring the Guest Roles and VAP Profile
Guest usage in enterprise wireless networks requires the following special consideration:
 Guest users must be separated from employee users by VLANs in the network.
 Guests must be limited not only in where they may go, but also by what network protocols and
ports they may use to access resources.
 Guests should be allowed to access only the local resources that are required for IP
connectivity. These resources include DHCP and possibly DNS if an outside DNS server is not
available. In most cases, a public DNS is always available.
 All other internal resources should be off limits for the guest. This restriction is achieved usually
by denying any internal address space to the guest user.


A time-of-day restriction policy should be used to allow guests to access the network only during
normal working hours, because they should be using the network only while conducting official
business. Accounts should be set to expire when their local work is completed, typically at the
end of each business day.

Access controlled

No access
after hours
arun_060

Figure 52


Guest access has a time limit

A rate limit can be put on each guest user to keep the user from using up the limited wireless
bandwidth. Employee users should always have first priority to the wireless medium for
conducting company business. Remember to leave enough bandwidth to keep the system
usable by guests. Aruba recommends a minimum of 10% of total bandwidth be made available
to guests. Guests can always burst when the medium is idle. For information about how to
configure these bandwidth parameters, see Traffic Management Profile on page 105.
Mobility
controller

Data

Controlled
data
arun_061

Figure 53

Guest access has a bandwidth limit
Configuring the Guest Roles and VAP Profile | 75



Unlike employees, the guest users typically log in through a captive portal. Usually, guests are
assigned two different roles. One role is assigned when they associate to the guest SSID and the other
is assigned when they authenticate successfully through the captive portal. Only the guests who
successfully authenticate are allowed to use to the services needed to connect to the internet.
The example network uses the guest-logon role as the initial role and the auth-guest role for
authenticated guests. Before you configure these two roles, first create the policies that are associated
with them.
The guest-logon role uses these policies:
 Amigopod
 captiveportal (predefined policy)
 guest-logon-access
 block-internal-access
The auth-guest role uses these policies:





guest-logon-access
block-internal-access
auth-guest-access
drop-and-log

Guest authentication and management can be provided through the internal resources of the Aruba
controller or through Amigopod. The internal resources of the Aruba controller can be used for visitor
management in small deployments. However, Aruba recommends the use of Amigopod for visitor
management in large campuses. For information on deploying the Aruba controller for visitor
management, see the ArubaOS 6.1 User Guide available on the Aruba support site. For more
information on the special features and deployment scenarios of Amigopod, see the Amigopod
deployment guide available at Amigopod support site. This VRD explains only the configurations
required on the Aruba controller when Amigopod is used for visitor management.

Configuring the Amigopod Policy
The Amigopod policy allows HTTP and HTTPS traffic only to the Amigopod server that is defined in the
Amigopod alias. This policy used in the preauthenticated role allows the client-based HTTP and
HTTPS traffic to reach the hosted captive portal pages on the Amigopod appliance.
Table 20 summarizes the rules used by the Amigopod policy.
Table 20

Rules Used by the Amigopod Policy

Rule
Source
Number

Destination

Service

Time Range

Action

Purpose

1

User

Alias
Amigopod

Service
svc-http

Working-hours

Scr-nat

This rule allows HTTP traffic from
the users to Amigopod server. The
permitted traffic is source-NATed.

2

User

Alias
Amigopod

Service
svc-https

Working-hours

Scr-nat

This rule allows HTTPS traffic from
the users to Amigopod server. The





CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session Amigopod
user
alias Amigopod svc-http src-nat
user
alias Amigopod svc-https src-nat
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 54

Amigopod policy

Configuring the guest-logon-access Policy
The guest-logon-access policy is similar to the predefined logon-control policy, but it is much more
restrictive. The guest-logon-access policy is a part of the guest-logon and auth-guest roles. The rules
defined in this policy allow these exchanges:
 Allow DHCP exchanges between the user and the DHCP server during business hours, but
block other users from responding to DHCP requests.
 Allow DNS exchanges between the user and the public DNS server during business hours.
Traffic is source-NATed using the IP interface of the controller for the guest VLAN.
Guest users are denied access to the internal network, so the Public-DNS alias is used. All the DNS
queries of the guest users are forwarded to these public DNS servers.
A time range is used to allow users to associate to the guest network only during certain hours of the
day. A time range called Working-hours is created and used in the example network.





Table 21 summarizes the rules used by the guest-logon-access policy.
Table 21

Rules Used by the guest-logon-access Policy

Rule
Number

Source

Destination

Service

1

User

Any

UDP
min port = 68
max port = 68

2

Any

Any

Service
svc-dhcp
(udp 67 68)

3

Any

Alias
Public-DNS

Service
svc-dns
(udp 53)

Time
Range

Action Purpose
Drop

This rule drops responses from a
personal DHCP server. This action
prevents the clients from acting as
DHCP servers. (This rule should
be active always and not just
during the working hours.)

Working-hours

Permit

This rule allows clients to request
and discover DHCP IP addresses
over the network. The DHCP
server on the network does not fall
under the user category.
Therefore, its response on port 68
is not dropped by the first rule. The
first two rules guarantee that
DHCP is processed only by
legitimate DHCP servers on the
network.

Working-hours

Scr-nat

This rule allows DNS queries only
to the DNS servers that are
defined in the Public-DNS alias.
The allowed traffic is sourceNATed.

CLI Configuration
MC1-Sunnyvale-3600
!
time-range "Working-hours" periodic Weekday 07:30 to 17:30
!
ip access-list session guest-logon-access
user any udp 68 deny
any any svc-dhcp permit time-range Working-hours
user alias Public-DNS svc-dns src-nat time-range Working-hours
!





WebUI Screenshot
MC1-Sunnyvale-3600

Figure 55

Figure 56


Time range

guest-logon-access policy




Configuring the block-internal-access Policy for the Guest Role
The internal resources of an organization should be available only to employees or to the trusted
groups. Guest users are not part of the trusted entity, so they must be denied access to all internal
resources. As the name implies, the block-internal-access policy denies access to all internal
resources. This policy is a part of the guest-logon and auth-guest roles.
Table 22 summarizes the rule used by the block-internal-access policy.
Table 22

Rule Used by the block-internal-access Policy

Rule
Number

Source

Destination

Service

Action

Purpose

1

User

Alias
Internal-Network

Any

Drop

This rule denies access to all the addresses
that are in the Internal- Network alias.

CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session block-internal-access
user alias Internal-Network any deny
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 57


block-internal-access policy




Configuring the auth-guest-access Policy
The most important purpose of the auth-guest-access policy is to define the protocols and ports that
the users are allowed to access. This policy is an integral part of the auth-guest role. The auth-guestaccess policy allows HTTP and HTTPS traffic to go to any destination from the user during business
hours. The traffic is source-NATed using the IP interface of the controller for the guest VLAN.
Table 23 summarizes the rules used by the auth-guest-access policy.
Table 23

Rules Used by the auth-guest-access Policy

Rule
Number

Source

Destination

Service

Time
Range

Action

Purpose

1

User

Any

Service
svc-http

Working-hours

Scr-nat

This rule allows HTTP traffic from
the users to any destination. The

2

User

Any

Service
svc-https

Working-hours

Scr-nat

This rule allows HTTPS traffic
from the users to any destination.
The permitted traffic is sourceNATed.

CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session auth-guest-access
user any svc-http src-nat time-range Working-hours
user any svc-https src-nat time-range Working-hours
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 58


auth-guest-access policy




Configuring the drop-and-log Policy
The drop-and-log policy denies all traffic and records the network access attempt.
Table 24 summarizes the rule used by the drop-and-log policy.
Table 24

Rule Used by the drop-and-log Policy

Rule
Number

Source

Destination

Service

Action

Log

Purpose

1

User

Any

Any

Deny

Yes

This rule denies access to all services on
the network and logs the network access
attempt.

CLI Configuration
MC1-Sunnyvale-3600
!
ip access-list session drop-and-log
user any any deny log
!

WebUI Screenshot
MC1-Sunnyvale-3600

Figure 59


drop-and-log




Configuring the Initial Guest Role
The guest-logon role is the first role that is assigned to the users when they associate with the guest
SSID. A user in this role has access only to the DHCP and DNS services. Unlike 802.1X/EAP, captive
portal is a Layer 3 type authentication. A user who associates to the guest SSID is given an IP address
and related DNS information even before he authenticates himself. When this user opens the browser
and tries to access a web page, the guest-logon role directs him to a captive portal page. The captive
portal page requires login credentials. The captive portal authentication profile that is appended to this
role specifies the captive portal login page and other configurable parameters, such as the default role,
the authentication server, and the welcome page. To create and add the captive portal authentication
profile to this initial guest role, see Configuring the Captive Portal Authentication Profile for Guest
WLAN on page 89.
Table 25 summarizes the policies used in the guest-logon role.
Table 25

Policies Used in the guest-logon Role

Policy
Number

Policy Name

Purpose

1

Amigopod

Allows the client-based HTTP and HTTPS traffic to reach the hosted captive
portal pages on the Amigopod appliance. If this policy is not used in the guestlogon role, the guest users cannot proceed to the login page on the Amigopod.
The preauthenticated guest logon policy is usually designed to deny all traffic
other than DHCP and DNS traffic. For details, see Configuring the Amigopod
Policy on page 76.

2

captiveportal (predefined policy)

This predefined policy initiates captive portal authentication. This policy redirects
any HTTP or HTTPS traffic to port 8080, 8081, or 8088 of the controller. When the
controller sees traffic on these ports, it checks the captive portal authentication
profile that is associated with the current role of the user and processes the
values specified on this profile.

3

guest-logon-access

This policy allows DHCP and DNS services. For details, see Configuring the
guest-logon-access Policy on page 77.

4


This policy blocks access to the entire internal network. For details, see
Configuring the block-internal-access Policy for the Guest Role on page 80.

CLI Configuration
MC1-Sunnyvale-3600
!
user-role guest-logon
access-list session Amigopod
access-list session captiveportal
access-list session guest-logon-access
access-list session block-internal-access
!





WebUI Screenshot
MC1-Sunnyvale-3600

Figure 60


guest-logon role




Configuring the Authenticated Guest Role
The auth-guest role is the role that is assigned to guest users after they authenticate successfully
through the captive portal. This role is the default role in the captive portal authentication profile. In
addition to restricting the network access to business hours, this role allows only HTTP and HTTPS
services to access the Internet.
If an organization wants its guest users to use the printers in the internal network, a separate policy
must be created that allows user traffic to an alias called printers. This alias must include only the IP
address of the printers that the guests are allowed to use. Place this policy in the auth-guest user role
just above the block-internal-access policy.
Table 26 summarizes the policies used in the auth-guest role.
Table 26

Policies Used in the auth-guest Role

Policy
Number

Policy Name

Purpose

1

cplogout (predefined policy)

This policy makes the controller present captive portal logout window. If the user
attempts to connect to the controller on the standard HTTPS port (443) the client
will be NATed to port 8081, where the captive portal server will answer. If this rule
is not present, a wireless client may be able to access the controller's
administrative interface.

2

guest-logon-access

This policy denies personal DHCP servers and provides legitimate DHCP
services and DNS.

3


This policy blocks access to internal network. This policy should be placed before
the next policy that allows HTTP and HTTPS service, otherwise guest users will
have access to the internal websites.

4

auth-guest-access

This policy allows HTTP and HTTPS services to any destination.

5

drop-and-log

Any traffic that does not match the previous policies encounters this policy. This
policy denies all services and logs the network access attempt.

CLI Configuration
MC1-Sunnyvale-3600
!
user-role auth-guest
max-sessions 65535
access-list session
access-list session
access-list session
access-list session
access-list session
!


cplogout
guest-logon-access
auth-guest-access
drop-and-log


Aruba Campus Wireless Networks

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (20)

En vedette

En vedette (20)

Similaire à Aruba Campus Wireless Networks

Similaire à Aruba Campus Wireless Networks (20)

Plus de Content Rules, Inc.

Plus de Content Rules, Inc. (20)

Dernier

Dernier (20)

Aruba Campus Wireless Networks